Excessive lateral patellar translation on axial computed tomography indicates positive patellar J sign.

PURPOSE: The purpose of the study was to quantify the patellar J sign using traditional computed tomography (CT) scans. METHODS: Fifty-three patients (fifty-three knees) who suffered from recurrent patellar instability were included and analyzed. The patellar J sign was evaluated pre-operatively during active knee flexion and extension. It was defined as positive when there was obvious lateral patellar translation, and negative when there was not. The CT scans were performed in all patients with full knee extension; and the parameters including bisect offset index (BOI), patellar-trochlear-groove (PTG) distance, and patellar lateral tilt angle (PLTA) were measured on the axial slices. All the three parameters were compared between the J sign-positive group (study group) and the J sign-negative group (control group). In addition, the optimal thresholds of the three CT scan parameters for predicting the positive patellar J sign were determined with receiver operating characteristic (ROC) curves, and the diagnostic values were assessed by the area under the curve (AUC). RESULT: Among the fifty-three patients (fifty-three knees), thirty-seven (70%) showed obvious lateral patellar translation, which were defined as positive J sign (study group), and the remaining sixteen (30%) who showed no lateral translation were defined as negative J sign (control group). The mean values of the three CT parameters in the study group were all significantly larger compared to the control group, including BOI (121 ± 28% vs 88 ± 12%, P = 0.038), PTG distance (5.2 ± 6.6 mm vs - 4.4 ± 5.2 mm, P < 0.05), and PLTA (34.9 ± 10.5° vs 25.7 ± 3.4°, P = 0.001). Furthermore, the evaluation of ROC analysis showed that the AUC of BOI was the largest (AUC = 0.906) among the three parameters, and the optimal threshold of BOI to predict the positive patellar J sign was 97.5% (Sensitivity = 83.3%, Specificity = 87.5%). CONCLUSIONS: In this study, the prevalence of positive patellar J sign was 70%. The BOI measured from the axial CT scans of the knee joint can be used as an appropriate predictor to differentiate the positive J sign from the negative J sign, highlighting that the excessive lateral patellar translation on axial CT scan indicates positive patellar J sign. LEVEL OF EVIDENCE: IV.